Research On Key Technologies Of Labeling And Querying Dynamic Xml Document

Along with the rapid development of network applications, XML has become a de facto standard of network data storage and exchange. More and more information processing system uses XML document as a carrier of information storage, exchange and posted. how to effectively manage XML data is a hot spot in academia now. One of the key problems of XML data manage is supporting XML data update and supporting efficient query processing. Aiming at exiting labeling schemes, which either support efficient query but do not support data update, or support data update but do not support efficient query, this paper made further research. The main research contents are as follows.Firstly, aiming at Extended Dewey labeling which suporrts efficient query, but it does not support labeling update, this paper puts forward a dynamic Extended Dewey labeling(DED) based on Dynamic Finite State Transducer(DFST), which makes DED both support labeling own update, and avoid labeling losing effectiveness when DTD(Document Type Definition) changes. DED can completely support XML data update. Then, in view of the high space storage consumption of existing labeling schemes, this paper proposes an improving storage solution based on Huffman labeling scheme, which reduces the storage space consumption.Secondly, in view of the parseing method of ED is pure. This paper proposes an Twig query pattern matching algorithm(EDTS) based on Reversible Finite State Transducer (RFST). Through cost model, RFST chooses the best parsing strategy, support two-way parsing, which improves the parsing efficiency and querying performance.Finally, through the rich experimental results verify the effectiveness of DED code and the efficiency of the EDTS algorithm.